Development of a Multi-Class Dress Code Detection System Utilizing RoboFlow Object Detection Model v3

Dress code violations in educational institutions present ongoing challenges for administrators striving to maintain a balance between enforcing standards and fostering a sense of individuality among students. This study seeks to address these challenges by developing an innovative multiclass dress code detection system utilizing the RoboFlow Object Detection Model v3. The motivation for this study stems from the need for a systematic and efficient method to manage dress code policies while promoting inclusivity and self-expression within school environments. The proposed system categorizes dress codes into four classes: school uniforms, organizational shirts, civilian attire, and formal wear. The research methodology encompasses meticulous data collection from a dataset of 10,000 images, preprocessing steps to enhance image quality, data augmentation to improve model robustness, and model training using the advanced COCOn checkpoint for high-performance object detection. Key preprocessing techniques include resizing, contrast adjustment, and cropping, while augmentation processes introduce variations to improve adaptability. The evaluation metrics reveal exceptional performance, with a Mean Average Precision (mAP) of 99.5%, precision of 99.8%, and recall of 100.0%. This system demonstrates the ability to accurately and efficiently identify and classify dress codes, offering significant improvements over existing approaches. Comparative analysis highlights its superiority in terms of precision and reliability, making it a practical solution for schools aiming to streamline dress code enforcement. Future work will explore the inclusion of larger datasets and integration with emerging object detection technologies to further refine accuracy and expand applicability. This study sets a benchmark for utilizing AI in educational policy enforcement, contributing to a more structured and inclusive campus environment.